Your bill is your plan
A hard cap with no overage, on every plan. Free builds three automations and runs them for real, five times a month, with no card.
Reach your allowance and AI pauses until it resets, or you top up. There is no overage and no invoice you did not choose. Every run carries its own ceiling too, so a single automation can never drain the month.
Free
Build an automation, wire it to your tools, and let it do the work for real. Five runs, no card.
- Build up to 3 automations, from a sentence
- They take real actions: send, post, file, collect
- They run on a schedule, unattended
- Ready-made apps & teams, one click
- Claude Haiku
These take real actions and run on a schedule. Pro adds building your own apps and teams, and lifts the five-run limit.
Pro
It takes real actions and runs itself, unattended.
- Everything in Free, plus:
- Unlimited runs, with no five-run ceiling
- Up to 50 automations
- Build your own apps & teams, from a sentence
- Up to 25 apps & teams
- Claude Sonnet, a smarter model
- Integrations, memory, RAG & web
- Email support
Business
For small teams building together.
- Everything in Pro, plus:
- Up to 100 apps & teams
- Up to 500 automations
- Shared workspace + roles
- Claude Opus, the most powerful model
- Bring your own model key
- Priority email support
Enterprise
Org-scale automation with governance, security and onboarding for larger deployments.
Contact sales- Unlimited apps, teams & seats
- Any model, custom credit volume
- SSO & organization policies
- Security review & questionnaire support
- SLA & dedicated support
- Custom integrations & onboarding
A credit is one unit of AI usage; building and running your automations both draw from your monthly allowance. It's a hard cap with no overage. Reach it and AI pauses until it resets, or top up anytime, so you never get a surprise bill. Every run also has a per-run ceiling, so one automation can't drain your month. Human-in-the-loop approval and server-side credentials are on every plan.
Rather run it on your own machine? Claritty Studio is free and open source. Automations run locally with your own keys and no account. Schedules fire while the app is open, so unattended runs on always-on infrastructure are what a plan buys.
Questions & answers
Build up to three automations and actually run them, five times a month, with no credit card. They take real actions and run on a schedule, exactly as they would on a paid plan, and nothing acts without your approval. What free runs out of is runs, not capability. Building your own apps and teams is the paid line.
Pro lifts the five-run ceiling and lets you build your own apps and teams rather than only automations, on Claude Sonnet. Business adds your whole team with roles and approvals, the most powerful model (Claude Opus), bring-your-own-key, and more scale.
Each plan includes a monthly credit allowance (metered in AI usage). It is a hard cap with no overage: when you reach it, AI pauses until your allowance resets next cycle, or you can top up, so you are never billed beyond your plan. Every automated run also has a hard per-run ceiling, so a single trigger or a bad prompt can never quietly drain your month. Allowances renew monthly and don't roll over.
No. Claritty runs the model for you and draws it from your plan's monthly allowance, there is no Anthropic or OpenAI key to manage. Business and Enterprise can bring their own key, and those calls run on your key instead.
Yes. Upgrade and the higher limits apply immediately; downgrade and it takes effect at the end of your billing period. We also offer a 30-day money-back guarantee on paid plans.
Your logins are brokered server-side and never touch app code, each automation runs isolated, and nothing irreversible happens without your approval, on every plan. See the security page for the details.
Run it for real, free
Build your first automation and let it act on your own data, five times a month, with no card. Upgrade when you want it running without a limit.
Start free